Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <1353c02a8a8f218aa795e5a1b317382f@smtp.hushmail.com>
Date: Fri, 14 Jun 2019 10:47:12 +0200
From: magnum <john.magnum@...hmail.com>
To: john-users@...ts.openwall.com
Subject: Re: Unable to use OpenCL on arm64v8/debian stretch

On 2019-06-14 06:46, billwhittaker@...e-mail.net wrote:
> 	As directed, I have compiled 1.9.0-jumbo-1 from openwall.
> 	I see no change, but do have additional information.

> ../run/john --test --format=wpapsk-opencl
> Device 1@...i: Mali-T860
> Benchmarking: wpapsk-opencl, WPA/WPA2/PMF/PMKID PSK [PBKDF2-SHA1 OpenCL]... 0: OpenCL CL_INVALID_PROGRAM_EXECUTABLE (-45) error in opencl_wpapsk_fmt_plug.c:270 - Error creating kernel

That's a strange error code, but anyway I'd expect it to be a 
driver/runtime bug. We see them a lot, from all vendors but nvidia. 
Somehow nvidia manage to end up with very few bugs compared to most 
everyone else including big guys like Intel and AMD.

> ../run/john --test --format=opencl
> Device 1@...i: Mali-T860
> Benchmarking: sha1crypt-opencl, (NetBSD) [PBKDF1-SHA1 OpenCL 4x]... DONE
> Speed for cost 1 (iteration count) of 64000 and 40000
> Raw:	12.5 c/s real, 2560 c/s virtual

This shows that *some* formats does work fine, so your OpenCL 
installation isn't totally broken.

That benchmark speed is pretty low though, my laptop GPU does 10510 c/s 
with that format. What if you add --force-scalar to the command line? By 
default, some formats will use 4x vectorizing because your runtime 
claims that's the preferred width of int, but it doesn't always end up 
faster.

Anyway I'd start with looking for OpenCL driver/runtimer updates unless 
you already did so.

magnum

Powered by blists - more mailing lists

Confused about mailing lists and their use? Read about mailing lists on Wikipedia and check out these guidelines on proper formatting of your messages.